Heart diseases are a major cause of mortality and morbidity. The high mortality rate and in survivors the high rate of disabilities is a social and economic burden on the Canadian economy. A faster detection of these life-threatening events and an earlier start of the therapy would save many lives and reduce successive handicaps. There is an urgent clinical need for (i) early detection and automatic classification of cardiac disease, (ii) the efficient monitoring of heart disease patients to provide warning if the state of the heart patient changes.
The heart attack analysis dataset is used in this study. The dataset contains information about 304 patients and 13 risk factors and heart attack risk output. The output response is the heart stroke risk presented by 1 and 0, there is risk and no risk respectively. The 13 risk factors used for the prediction of heart attack risk are as follows: age, gender, chest pain type, resting blood pressure, cholesterol, fasting blood sugar, resting electrocardiographic result, maximum heart rate achieved, exercise-induced angina, ST depression induced by exercise relative to rest, number of major vessels, thalassemia.
Through this research project, we created a system that can help clinicians significantly in predicting stroke risk in patients and also monitor it. We created a web application that can employ the recommender system machine learning model that we created to predict heart stroke risk levels and rank them and display them to the users in an interactive interface. The system also generates a clinical PDF report to aid the clinician. All the patient data is stored on the system in a secure database hence ensuring ease and availability for clinicians to access patient records in the future.